(19)
(11) EP 4 390 772 A3

(12) EUROPEAN PATENT APPLICATION

(88) Date of publication A3:
28.08.2024 Bulletin 2024/35

(43) Date of publication A2:
26.06.2024 Bulletin 2024/26

(21) Application number: 24175389.6

(22) Date of filing: 18.03.2020
(51) International Patent Classification (IPC): 
G01L 5/00(2006.01)
G01L 5/22(2006.01)
G06N 3/08(2023.01)
 B25J 9/16(2006.01)
G05B 13/02(2006.01)
(52) Cooperative Patent Classification (CPC):
G01L 5/009; G01L 5/226; G06N 3/063; G06N 3/008; B25J 9/1612; B25J 9/1694; B25J 9/1692; G05B 2219/39058; G05B 2219/32192; G05B 2219/40625; G05B 2219/39532; G06N 3/045
(84) Designated Contracting States:
AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

(30) Priority: 19.03.2019 US 201916358485

(62) Application number of the earlier application in accordance with Art. 76 EPC:
20163869.9 / 3712584

(71) Applicant: NVIDIA Corporation
Santa Clara, CA 95051 (US)

(72) Inventors:
  • Birchfield, Stan
    Santa Clara, CA, 95051 (US)
  • Boots, Byron
    Santa Clara, CA, 95051 (US)
  • Fox, Dieter
    Santa Clara, CA, 95051 (US)
  • Handa, Ankur
    Santa Clara, CA, 95051 (US)
  • Ratliff, Nathan
    Santa Clara, CA, 95051 (US)
  • Sundaralingam, Balakumar
    Santa Clara, CA, 95051 (US)
  • Lambert, Alexander
    Santa Clara, CA, 95051 (US)

(74) Representative: Kraus & Lederer PartGmbB 
Thomas-Wimmer-Ring 15
80539 München
80539 München (DE)

   


(54) FORCE ESTIMATION USING DEEP LEARNING


(57) A computer system generates a tactile force model for a tactile force sensor by performing a number of calibration tasks. In various embodiments, the calibration tasks include pressing the tactile force sensor while the tactile force sensor is attached to a pressure gauge, interacting with a ball, and pushing an object along a planar surface. Data collected from these calibration tasks is used to train a neural network. The resulting tactile force model allows the computer system to convert signals received from the tactile force sensor into a force magnitude and direction with greater accuracy than conventional methods. In an embodiment, force on the tactile force sensor is inferred by interacting with an object, determining the motion of the object, and estimating the forces on the object based on a physical model of the object.







Search report






Search report